The 2-year survival of patients with high-grade, primary brain tumors is only 30% with only a small further decline at 10 years to 20%. These abysmal survival figures, unchanged over the last 30 years, suggest effective local tumor control is not established during treatment. This failure is not detected by current imaging methods until months after completing treatment. Thus, vital time is lost for these patients before effective therapy can be instituted. More aggressive early local control may greatly improve treatment success if targeted therapy can be balanced against adverse side effects. Novel sodium imaging biomarkers are to be investigated for sensitivity to short- term tumor responses and for predictive value for recurrence as a first step towards customized radiation treatment of brain tumors. Regional changes in density of tumors that occur during fractionated external beam radiation treatment of primary brain tumors in humans are reflected in the changes in tissue sodium concentration, as can be measured using quantitative sodium magnetic resonance imaging. Maps of this sodium biomarker, reflecting early tumor response during fractionated radiation treatment, are to be correlated on a regional basis to the accumulated radiation dose and to tumor recurrence within a 2-year follow-up period. The goal is to determine regional sensitivity of early sodium response for predicting tumor recurrence under current standards of radiation treatment. This biomarker could be used to guide adaptive radiation therapy during initial treatment specifically to predicted areas of treatment failure. Radiobiologic models of the temporal trends in these new biomarkers will be used to estimate the radiation sensitivity parameters for tumors for the first time in individual patients. These parameters may explain regional treatment failures in terms of non- uniform radiation sensitivity in a uniform radiation treatment field. If these early changes in the sodium imaging biomarker can identify areas of ultimate treatment failure, they offer potential to tailor radiation therapy for individual patients to improve outcome. Large treatment trials can be avoided. This imaging approach is an important step on the NIH roadmap towards personalized healthcare.